|
|
| |||||||||||||||||||||||||||||||||||||||||
Pay for performance is one strategy for moving from payment based solely on the quantity of services rendered to payment based on the quality or efficiency of care.2 Such payment systems are now widely used in private health plans, and Medicare is gaining experience with different pay-for-performance designs.3,4,5 Most designs reward clinically high-quality care or patient-centered care; few reward care coordination or increased efficiency over time in the treatment of a particular condition.3
One way to reward care coordination and efficiency is to base payment, in whole or in part, on the total care of a patient during an acute episode of illness or during a period of time. Pure capitation payment, which puts providers at financial risk, is one example of this approach, but it did not succeed in the mid-1990s, owing to backlash from patients about the managing of costs by managed-care plans and from physicians about the financial risks.
An alternative is to reduce variation in payments for acute episodes of illness or for longitudinal care of patients with chronic conditions. For example, in the Medicare program in 2002, adjusted charges for the care of patients with acute myocardial infarction ranged from $20,720 in Knoxville, Tennessee, to $47,133 in New York City.6 Among regional groups of Medicare beneficiaries with all three chronic conditions — diabetes, chronic obstructive pulmonary disease, and congestive heart failure — annual adjusted charges were at least twice as high in the highest-cost groups as in the lowest-cost groups.7 A global fee for "care episodes" — the total cost of hospital services, physician services, and other services required for treating an acute condition or the total cost for all the care required during a given year for a patient with chronic conditions — with appropriate adjustment for complexity of the case mix would reward providers who have lower costs while penalizing higher-cost providers. Extra payments to teaching hospitals or other providers for additional services, such as teaching, research, or care of the uninsured, would be supplemental to the global fee.
The major issue in designing payment systems for care episodes is assigning accountability for care across different settings and over time. In this issue of the Journal, Pham et al. describe the care patterns for 1.79 million Medicare beneficiaries and show that care is typically highly dispersed among many physicians.8 Even when a patient was assigned to the physician who billed for the greatest number of that patient's evaluation and management visits, that physician accounted for only 35% of the patient's total visits. The average patient saw two primary care physicians and five specialists, working in a median of four practices, over the course of a given year. For patients with a chronic condition, such as diabetes, coronary artery disease, or lung cancer, an even greater number of physicians and practices were involved in care. These findings highlight both the need to improve the longitudinal care of Medicare beneficiaries and the difficulty of attributing accountability for care to a given physician or set of physicians.
Pham et al. also found that the assigned physician changed from one year to another for one third of beneficiaries. During the period from 2000 to 2002, almost half the beneficiaries (46%) were assigned a new physician. Having a regular source of care and continuous care with the same physician over time have been associated with better health outcomes and lower total costs.9 The cost of care and rates of medical errors, by contrast, are greater when patients are cared for by many physicians.6,10,11
The challenge is not just to design a pay-for-performance system that can rationally assign care to the various physicians involved in treating Medicare beneficiaries. It is also to change current practices to ensure a stable and responsible primary source of care and to improve care coordination. Pham et al. suggest the possibility of prospective designation of physicians responsible for a patient's care. The Institute of Medicine Committee on Redesigning Health Insurance Performance Measures, Payment, and Performance Improvement Programs, on which I served, endorsed this strategy. The committee recommended that Medicare encourage beneficiaries and providers to identify providers to serve as the principal sources of care and to reward these providers for care coordination that meets specified standards.1 The American College of Physicians recently proposed Medicare demonstrations of the "patient-centered medical home" concept (in which physicians and practices provide care that is accessible, centered around patients and families, comprehensive and continuous, coordinated, equitable, and culturally sensitive) and suggested responsibilities of physician practices to be used, including ensuring accessibility of care, care coordination, and providing high-quality care.12 Congress provided for a demonstration in the Tax Relief and Health Care Act of 2006 to test whether paying such physician practices for assuming all the responsibilities of a patient-centered medical home results in lower total Medicare costs.
Another strategy would be to encourage the growth of integrated delivery systems, large physician group practices, or networks of physician practices and to encourage these entities to assume responsibility for the total care of patients by allowing them to participate in Medicare and by designing payment systems that reward both the quality and efficiency of care. The ongoing Medicare physician group practice demonstration is one example of this approach; it rewards group practices that meet quality targets yet whose costs for Medicare beneficiaries increase less over time (by at least two percentage points) than the costs for similar beneficiaries at other practices.4
In the Hospital Quality Incentive Demonstration of the Centers for Medicare and Medicaid Services and Premier (a company that provides services to hospitals), hospitals — but not individual physicians — receive bonus payments.5 To build on this model, physicians involved in the care of a patient in the hospital could be eligible for bonuses if their hospital received bonuses for such care. Bonuses for the quality of care could have the condition of meeting efficiency thresholds based on the total longitudinal cost of care. Alternatively, payments to hospitals and physicians for care episodes could be based on the performance of all medical staff affiliated primarily with a particular hospital.13
Yet pay for performance is unlikely to fundamentally alter the incentives in the fee-for-service payment system. Ideally, it would serve as an interim program in the transition to fundamental payment reform. Ultimately, the payment of primary care physicians might be a blend of fee for service, monthly fees for practices serving as patient-centered medical homes, and additional bonuses for meeting quality and efficiency performance goals.14 Primary care physicians could instead be paid a monthly fee per patient (adjusted for the case mix), along with a bonus for performance.12,15 Such payment methods are more likely to be acceptable to physicians than is global capitation, because they do not put physicians at risk for services provided by others, and they contain only positive performance bonuses.
Given the dispersion of care across physicians and practices, fragmentation of the health system, and lack of continuity in physician–patient relationships, extensive evaluations of these new payment methods are warranted. A cautious move toward a more rational payment system, with opportunities to fine-tune payment as experience is gained, shows the greatest promise of avoiding unintended consequences. This caution, however, must be balanced against the urgent need to address the financial stresses from ever-rising costs of health care.
No potential conflict of interest relevant to this article was reported.
Source Information
From the Commonwealth Fund, New York.
References
| |||||||||||||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved. |
Previous
